In astrophotography, it is often difficult to obtain an image of a more faint object, such as a nebula, a spiral arm of a distant galaxy, or a distant star, when there is a closer and brighter star along an immediately adjacent line of sight. The difficulty is that in trying to get a longer exposure time for the more faint object (e.g., to increase the signal-to-noise ratio), such longer exposure of the adjacent brighter star within the field of view often overexposes or saturates the area where the bright star is located and may cause bleeding or streaking into the immediate surrounding areas.
The recent use of digital image sensors, such as a charge coupled device (CCD), as well as the recent improvements in increased resolution (more pixels or sensors per square inch) and increased sensitivity (higher signal-to-noise ratio) for CCDs, has made it possible to obtain better images of faint objects. However, the problem of obtaining a good image of a more faint object adjacent to a bright object still persists. Not every photon that strikes a CCD element actually produces a count (adding another electron to the charge). Also due to the quantum nature of light, this hit-or-miss process of detecting light is somewhat random. Noise occurring within a CCD imaging process also occurs at a random rate. The noise can come from a variety of sources: readout noise, dark count noise, background noise, and processing noise. However, even in light of the inherent randomness of CCD light sensing, the desired signal (e.g., light from a celestial object) repeats at a random rate that is exponentially larger than the random rate of noise. In other words, noise does not contribute or register as many photon counts at a CCD element as fast as signals from an image or light source. Therefore as a general rule in CCD imaging, longer exposure times markedly improve the signal-to-noise ratio, and hence improve the image quality.
There are techniques for reducing readout noise in a CCD array to increase the signal-to-noise ratio (to improve the image quality and contrast), such as binning Binning is a process of combining adjacent pixels of a CCD into groups of “superpixels,” such as two-by-two blocks of pixels or three-by-three blocks of pixels grouped together to form larger superpixels. Although binning may increase sensitivity, it does so at a sacrifice to resolution because binning effectively decreases the number of pixels per square inch. Hence, there is a need for a way to increase sensitivity without reducing resolution.
Another technique used in CCD imaging is antiblooming. As mentioned above, the problem that often arises with longer exposure times is saturation at areas within the field of view where a bright object floods the CCD elements with photons. Excess charge or buildup of electrons will sometimes bleed across a row or down a column of pixels causing an unwanted streak across the recorded image. Antiblooming tries to prevent such bleeding or streaking by diverting excess charge generated in the photosites to an antiblooming sink instead of the shift registers. Hence, the antiblooming technique acts as a clipping circuit for the CCD output. However, antiblooming may produce side effects like increased dark current (or dark count noise) and reduced sensitivity. Dark current is the accumulation of electrons at a CCD element, even in the absence of light, that produces a signal indistinguishable from one produced by light. Thus, a need exists for a way to handle such over-saturation events (e.g., bright objects in the field of view) that call for the use of antiblooming techniques, but without increased dark current and reduced sensitivity.
There are also other techniques for cleaning up an image after obtaining the image, such as digitally enhancing an image with a software application (e.g., Photoshop), tricolor imaging, or stacking images. But, such techniques can be time consuming and labor intensive. Also, such post-capture image processing challenges the integrity of the images. Hence, there is a need for a way to obtain or capture a better image in the first place before attempting to touch-up the image after it is recorded.
The technique of stacking images combined with other of the above techniques is currently one of the best known ways of generating an image or photograph that provides a high-quality image of faint objects and bright objects adjacent to one another. For example in stacking images, a first image may be taken from a first field of view that does not include the bright object or that only catches a small part of the bright object in a corner or edge of the image. This first image may be taken with a longer exposure time to obtain a better signal-to-noise ratio for the faint objects, and then the remnants of the bright image may be digitally cropped or deleted in post-image-capture processing. Then a second image with a different field of view including the bright object therein is obtained at a much shorter exposure time, which will likely result in a lot of background noise and very little, if any, capturing of the faint objects. Then the second image may be cropped and digitally enhanced with software to remove the background noise in post-image-capture processing. Next, the processed second image may be stacked with or onto the first image to provide a complete image for an artificial field of view including the bright and faint objects. Such stacking technique may require multiple layers and multiple images combined to obtain a comprehensive image containing bright and faint objects. However, again, such stacking may be quite time consuming and labor intensive, and the resulting image may not be proportional or to scale in the relationship among different portions of the image. Hence, this technique does not allow the desired image to be captured in a single field of view. Therefore, a need exists for a way to capture and record a high-quality image from a single field of view containing both bright and faint objects.